A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Liangchen Li, Jiajun He
Bidirectional consistency models extend consistency models to enable both forward and reverse sampling, improving image generation quality and flexibility.
Unknown
This paper presents a systematic review of bidirectional human-AI alignment, offering clarifications, a framework, and future research directions.
Pere Martra, Eugenio Martínez Cámara, Alfonso Ureña López
Fairness Pruning introduces a lightweight structural intervention that locates demographic bias in GLU-MLP layers by zeroing at most 40 neurons, achieving 99.49% capability retention while destabilizing bias bidirectionally.
Unknown
NV-Embed introduces a latent attention layer and bidirectional attention to significantly boost LLM performance on general-purpose text embedding tasks.
Unknown
NeoBERT redefines bidirectional encoders with optimal depth-to-width ratio, 4K context, and 250M parameters, achieving state-of-the-art GLUE and MTEB scores.
Unknown
UniLM uses a shared Transformer with custom attention masks to unify bidirectional, unidirectional, and sequence-to-sequence pre-training for both NLU and NLG.